ClawHub
Back to skill

Security audit

Fish Audio S2 Pro TTS

Security checks across malware telemetry and agentic risk

Overview

This is a documentation-only skill for running Fish Audio text-to-speech and voice cloning, with sensitive voice-data handling that is disclosed but needs careful use.

Install only if you are comfortable running external Fish Audio packages, containers, and model weights. Use an isolated environment, keep any API bound to localhost unless you add access controls, avoid untrusted ref_audio URLs or file:// inputs, only clone voices with explicit permission, and manually review or clean cached speaker files if voice profiles should not remain on disk.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill provides explicit voice-cloning examples using `ref_audio` and `ref_text` but omits any warning about consent, impersonation risk, biometric privacy, or lawful use of someone else's voice. In a TTS skill, this omission materially increases the chance of misuse because users are directly shown how to clone voices from supplied reference audio.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.